Sampling-free Uncertainty Estimation in Gated Recurrent Units with Applications to Normative Modeling in Neuroimaging

Seong Jae Hwang, Ronak R. Mehta, Hyunwoo J. Kim, Sterling C. Johnson, Vikas Singh, "Sampling-free Uncertainty Estimation in Gated Recurrent Units with Applications to Normative Modeling in Neuroimaging", UAI, 2019.

http://pitt.edu/~sjh95/

Code

See code/gru_predictor_controlled_mnist_train_v1.py

Data

1. Moving Digit MNIST from this paper is essentially built off of the traditional MNIST digits. Various properties of the sequences (e.g., directions, speed, sizes, etc.) can be adjust within the data generator.
2. Neuroimaging data (WRAP): The longitudinal neuroimaging data is a part of the Wisconsin Registry for Alzheimer's Prevention (WRAP) at the University of Wisconsin-Madison.